A phase I thorough QT/QTc study evaluating therapeutic and supratherapeutic doses of avacopan in healthy participants

Abstract

This phase I thorough QTc, double‐blind, randomized, placebo‐ and positive‐controlled, parallel group, multiple‐dose study evaluated avacopan's effect on cardiac repolarization using concentration‐QTc (C‐QTc) as the primary analysis. Avacopan 30 mg b.i.d. (therapeutic dose) was administered orally on days 1 through 7 followed by avacopan 100 mg b.i.d. (supratherapeutic dose) on days 8 through 14 in 29 healthy participants. Moxifloxacin 400 mg and placebo were administered on days 1 and 15 in a nested crossover design for assay sensitivity in separate cohorts to 28 participants. Time‐matched plasma concentrations and up to 10 replicate ECGs were obtained on prespecified days at baseline and postdose on days 1, 7, 14, and 15. The mean change from baseline on QTcF for avacopan (−5.5 to 3.5 ms) was similar to placebo (−6.9 to 1.4 ms) across days 1, 7, and 14. The mean effect on ΔΔQTcF (90% CI) was estimated as 1.5 ms (−0.17 to 3.09) and 0.8 ms (−2.41 to 4.05) for 30 and 100 mg avacopan b.i.d. treatments, respectively. Based on the C‐QTc analysis, avacopan's effect on ΔΔQTcF >10 ms can be excluded within the observed plasma concentration range of up to ~1220 and ~335 ng/mL for avacopan and active major metabolite, M1, respectively. The estimated population slopes showed a shallow relationship, which was not statistically significant. There was no clinically meaningful effect of avacopan on heart rate or cardiac conduction (PR and QRS intervals). Avacopan appeared to be generally well tolerated in this study population.

STUDY HIGHLIGHTS.

• WHAT IS THE CURRENT KNOWLEDGE ON THE TOPIC?

The effect of avacopan at therapeutic and supratherapeutic doses on QTc prolongation in healthy participants is unknown.

• WHAT QUESTION DID THIS STUDY ADDRESS?

Avacopan does not prolong QTc interval in healthy participants.

• WHAT DOES THIS STUDY ADD TO OUR KNOWLEDGE?

This study provides compelling evidence on the cardiovascular safety of avacopan at doses of 30 mg b.i.d. and 100 mg b.i.d.

• HOW MIGHT THIS CHANGE CLINICAL PHARMACOLOGY OR TRANSLATIONAL SCIENCE?

This study supports the cardiovascular safety of avacopan, which is a new treatment approved as an adjunctive treatment for adult patients with GPA and MPA.

INTRODUCTION

Avacopan (chemical name: (2R,3S)‐2‐[4(cyclopentylamino)phenyl]‐1‐(2‐fluoro‐6‐methylbenzoyl)‐N‐[4‐methyl‐3‐(trifluoromethyl)phenyl]piperidine‐3‐carboxamide; formerly known as CCX168) is a highly potent and selective orally administered small molecule antagonist of the human complement 5a receptor (C5aR, also called CD88). It was approved at a dose of 30 mg twice daily (b.i.d.) with food in the United States in October 2021 as an oral adjunctive treatment in adult patients with severe active granulomatosis with polyangiitis (GPA) and microscopic polyangiitis (MPA) in combination with standard therapy including glucocorticoids. Avacopan acts through antagonism of the interaction between C5aR and the anaphylatoxin C5a; the latter is produced through activation of the complement cascade. As a result, avacopan reduces neutrophil activation, chemoattraction, and integrin expression. Avacopan inhibits vascular endothelial cell retraction and permeability, thus ameliorating the necrotizing vasculitis associated with GPA and MPA. Notably, avacopan acts very specifically at this terminal part of the complement cascade, leaving the rest of the complement system, including C3a and C3b intact. Through its specificity, avacopan does not block the production of C5b‐9 (membrane attack complex), thus leaving this host defense mechanism in place. ¹ , ² , ³

The pharmacokinetics (PK) of avacopan have been extensively evaluated following single and multiple doses in healthy participants and following twice daily doses for up to 1 year in phase II and III studies in patients with GPA and MPA. Following single oral doses of avacopan 1–100 mg and multiple doses ranging from 10 to 100 mg in the first‐in‐human study, the time to reach maximum concentration (T _max) was approximately 2 h, and avacopan exposures (area under the curve [AUC] and the maximum concentration [C _max]) increased dose proportionally across the dose range evaluated. ² Administration of avacopan with a high‐fat and high‐calorie meal resulted in an increase of ~70% in AUC and a delay in T _max, but no significant effect on C _max. Thus, avacopan is administered with food. ³ , ⁴ Avacopan is primarily metabolized by the cytochrome P450 (CYP) 3A4 enzyme, which results in the formation of a monohydroxylated product, M1, the major circulating metabolite, representing ~12% of the total drug‐related material in plasma and has approximately the same activity as avacopan on the C5aR. ³ The plasma protein binding of avacopan and metabolite M1 is greater than 99.9%. The main route of clearance of avacopan is phase I metabolism followed by biliary excretion of the metabolites into feces. The PK properties of avacopan are not meaningfully affected by hepatic function or renal function. ³

Per the International Council for Harmonization (ICH) E14 guidelines, new chemical entities in drug development must undergo a careful evaluation of potential effects on electrocardiogram (ECG) parameters, with emphasis on the QTc interval. ⁵ Drug‐induced QTc prolongation can lead to life‐threatening adverse events (AEs), such as syncope or death due to torsade de pointes. ⁶ Until recently, regulatory guidance has recommended the use of a by‐timepoint analysis as the statistical end point for a dedicated QT trial. However, the R3 Q&A document updating the ICH E14 guidance from 2015 specified that concentration‐response analysis can serve as an alternative to the by‐timepoint analysis or Intersection Union Test as the primary basis for decisions to classify the risk of a drug. ⁷

Previously, the first‐in‐human study did not identify any clinically meaningful impact of avacopan following single ascending doses or multiple repeated doses on QT/QTc prolongation. ² Although the results were informative, this study had limitations in that it did not include a positive control to assess assay sensitivity and the highest evaluated dosing regimen was 50 mg b.i.d. for 7 days, which did not provide adequate coverage for any expected increases due to extrinsic or intrinsic factors. In a later conducted food effect study, the impact of avacopan following a single dose at 30 mg or b.i.d. doses of 100 mg without food was evaluated on QTc prolongation using correction by Fridericia method (QTcF) (unpublished). Overall, there did not appear to be any relationship between avacopan concentrations and ECG parameters. However, these results still did not provide a definitive assessment of the impact of avacopan exposures on QTc prolongation due to non‐assessment of assay sensitivity. Therefore, a phase I thorough QT (TQT) study as described in this manuscript was conducted to meet the requirement as defined in the ICH E14 document. This was a multiple‐dose, randomized, double‐blind, placebo‐ and positive‐controlled, parallel‐group study evaluated the effect of avacopan at therapeutic (30 mg b.i.d.) and supratherapeutic (100 mg b.i.d.) doses on the QT/QTc prolongation in healthy participants. Moxifloxacin was used as a positive control to assess ECG assay sensitivity of the study. ⁸

METHODS

Study design and treatments

This phase I, double‐blind, randomized, placebo‐ and positive‐controlled, double‐dummy, parallel group, multiple‐dose study was conducted in healthy adult male and female participants at a single clinical site (Celerion, Tempe, Arizona, USA; NCT05988034). Moxifloxacin and placebo were combined into one treatment group with a nested crossover comparison between the positive control and placebo to demonstrate assay sensitivity.

Once eligibility was confirmed, participants were randomized 2:1:1 (Cohorts 1 [avacopan], 2A [moxifloxacin 400 mg/placebo], and 2B [placebo/moxifloxacin 400 mg]). Cohort 1 received multiple doses of avacopan (30 mg b.i.d. [60 mg total daily dose] for 7 days followed by 100 mg b.i.d. [200 mg total daily dose] for 7 days). Cohort 2 received placebo or 400 mg oral moxifloxacin in a nested crossover fashion. Participants were admitted to the clinical research unit on day −2, were dosed on day 1 through day 15, and were discharged on day 16 (Figure 1).

FIGURE 1.

Schema of the study design. *Intensive PK sampling was done on days 1, 7, 14, and 15. Only predose “trough” sampling was done on days 4, 5, 6, 11, 12, and 13. b.i.d., twice daily; D, day; ECG, electrocardiogram; PK, pharmacokinetics.

The primary end point was the mean change from baseline QTcF (ΔQTcF). Secondary end points included placebo‐corrected change from baseline in heart rate (HR), QTcF, PR, and QRS (∆∆HR, ∆∆QTcF, ∆∆PR, and ∆∆QRS) and the treatment‐emergent changes in T‐wave morphology and presence of U waves.

Avacopan was supplied in hard gelatin capsules, each of which contained 10 mg avacopan formulated in polyethylene glycol 4000 (PEG‐4000) and polyoxyl‐40 hydrogenated castor oil (Kolliphor® RH 40, formerly Cremophor® RH 40). Moxifloxacin for positive control was supplied as a 400 mg film‐coated tablet. Placebo to match avacopan or moxifloxacin was supplied as matching capsules or tablets as appropriate. The study drugs were administered orally following an overnight fast with ~240 mL of water at ambient temperature.

Study population

Participants included healthy male or female participants between the ages 18 and 60 years (inclusive) with body mass index (BMI) between 18 and 30 kg/m² (inclusive), body weight ≥50 kg and in good health, defined as having no clinically significant findings from medical history, physical exam, 12‐lead ECG, vital signs, and clinical laboratory evaluations. Key exclusion criteria included history or signs of cardiovascular disease, positive hepatitis B or C‐test, positive human immunodeficiency virus test, and known hypersensitivity to drugs, including moxifloxacin. As avacopan was primarily metabolized by CYP3A4, the study also excluded participants who were taking strong or moderate CYP3A4 inducers within 2 weeks prior to day 1 of avacopan treatment. ³

Compliance with ethical standards

This study was conducted in accordance with the ICH E6 Good Clinical Practice (GCP) protocol, with the ethical requirements referred to in the European Union (EU) directive 2001/20/EC and the ethical principles set forth in the Declaration of Helsinki. The clinical portion of this study was performed in compliance with Celerion Standard Operating Procedures. Prior to initiation of any study‐specific procedures, all pertinent study documents were reviewed by the Advarra (Columbia, MD, USA). The Institutional Review Board (IRB) and participants received a copy of the Informed Consent Form (ICF) that summarized the purpose of the study, the procedures to be carried out, and the potential hazards in non‐technical terms. All participants provided written informed consent before enrollment in the study and could withdraw at any time.

Electrocardiogram sample collection

Serial ECGs were extracted from continuous ECG recordings using a M12R continuous 12‐lead digital recorders (Global Instrumentation LLC, Manlius, New York, USA) supplied and supported by Clario (Philadelphia, Pennsylvania). Continuous Holter recordings were performed at prespecified timepoints on day −1 (baseline) and days 1, 7, 14, and 15 from predose through 24 h postdose. ECG extractions were time‐matched with PK timepoints. Up to 10 replicate digital 12‐lead ECGs were extracted at each timepoint at baseline (day −1) and on days 1, 7, 14, and 15 at the following timepoints: predose and 0.5, 1, 2, 3, 4, 5, 6, 9, 12, and 24 h postdose. ECGs were read centrally by Clario using a highly precise ECG technique (Early Precision QT). ECG intervals were measured by the core laboratory in a blinded manner.

Pharmacokinetic sample collection and analysis

Blood samples for the determination of plasma concentrations of avacopan and metabolite M1 were collected on days 1, 7, and 14, at predose and at 0.5, 1, 2, 3, 4, 5, 6, 9, 12, and 24 h post‐first dose on each day. In addition, trough PK samples were obtained at predose on days 4, 5, and 6 (corresponding to 12 h postdose from the second dose on days 3, 4, and 5, respectively) and predose on days 11, 12, and 13 (corresponding to 12 h postdose from the second dose on days 10, 11, and 12, respectively) following multiple doses of 30 mg or 100 mg avacopan, respectively. Blood samples for the determination of moxifloxacin levels were collected on days 1 and 15 at predose and 0.5, 1, 2, 3, 4, 5, 6, 9, 12, and 24 h postdose.

Plasma concentrations for avacopan, its metabolite M1, and moxifloxacin were determined using validated methods using liquid chromatography‐mass spectrometry/mass spectrometry through a validated method at Altasciences (Laval, Quebec, Canada). The calibration ranges were 1.00–500 ng/mL for avacopan and metabolite M1 and 10.0–4000 ng/mL for moxifloxacin.

Pharmacokinetic analysis

Plasma PK parameters for avacopan, metabolite M1, and moxifloxacin were calculated using standard noncompartmental methods with Phoenix® WinNonlin® Version 7.0 software (Certara). Descriptive statistics were also calculated for each PK parameter.

Primary analysis: concentration‐QTc analysis

The PK/QTc population included participants who had time‐matched QT/QTc and PK assessments. Before the C‐QTc analysis was performed, model assumptions were assessed, including assessment of drug effect on HR, relation between QTc and HR, time delay between concentrations and ΔQTc (hysteresis), and assessment of a linear C‐QTc relationship. ⁹ QTc prolongation using Fridericia method, also known as QTcF method, was used in this study. Other method for heart rate correction such as Bazett's QT correction (QTcB) was not evaluated as the limitations of this methodology have been widely acknowledged. ¹⁰ If a clinically relevant impact of avacopan on QTcF would have been observed, then subject‐specific QTcI would be evaluated. The relationship between plasma concentrations of avacopan and its metabolite M1 and ΔQTcF was quantified using a linear mixed‐effects modeling approach, which included ΔQTcF as the dependent variable, time‐matched plasma concentrations of avacopan, M1, or placebo as explanatory variates and time as fixed effect, a random intercept, and a slope as described in Supplemental Methods.

Overall, three models were explored: the full model with both analytes (avacopan and M1), a model with avacopan alone, and a model with M1 alone. The best model was selected based on the lowest Akaike Information Criterion (AIC). The adequacy of the model was further evaluated using goodness of fit plots. Model‐predicted effect and its two‐sided 90% confidence interval (CI) for ΔΔQTcF were obtained. If the upper bound of the two‐sided 90% CI as well as clinically meaningful plasma levels were below 10 ms, it would be concluded that neither avacopan nor M1 causes clinically concerning QTc prolongation within the observed range of plasma concentrations.

Assay sensitivity was assessed in a similar manner wherein if the slope of the C‐QTc was statistically significant at the 10% level for the two‐sided test and the lower bound of the two‐sided 90% CI of the predicted effect was above 5 ms at the geometric mean C _max for moxifloxacin, assay sensitivity would be deemed to have been demonstrated. The following baseline values were used for moxifloxacin: for participants in Cohort 2A, day 15 was used as the baseline, and for participants in Cohort 2B, day 1 was used as the baseline for moxifloxacin. The following baseline values were used for placebo participants: for participants in Cohort 2A, day −1 was used as the baseline, and for participants in Cohort 2B, day 14 was used as the baseline.

Model details are shown below:

$$\begin{array}{c}{\text{ΔQTcF}}_{\mathit{ijk}}=\alpha \times {\mathrm{TRT}}_j+{\beta }_k\times t_k+{\eta }_{\mu ,i}\hfill \\ +\left({\theta }_1+{\eta }_{\mathrm{c}1,\mathrm{i}}\right){\text{CONC1}}_{\mathit{ijk}}+\left({\mathrm{\theta }}_2+{\eta }_{\text{c2},i}\right){\text{CONC2}}_{\mathit{ijk}}+{\varepsilon }_{\mathit{ijk}},\hfill \end{array}$$

where i was the ith subject, j was the jth treatment, k was the kth timepoint, TRT_j was the jth treatment effect (active = 1 or placebo = 0), t _k was the kth time effect, CONC1 _ijk and CONC2 _ijk were the concentrations of avacopan and M1 at the kth timepoint for treatment j for subject i, respectively. α and β _k were the fixed effects for treatment and time, respectively. θ₁ and θ₂ were the fixed effects for the slopes of concentrations, η _μ,i , η_c1,i , and η_c2,i were the subject‐specific random effects for the intercept and slopes, respectively, having mean [0,0,0] and unstructured covariance matrix, and ɛ were independent residuals having mean zero and variance σ. ² The models for the two individual models included only one of the two analytes in the above equation.

By‐timepoint analyses

The analysis for QTcF for avacopan versus placebo was based on a linear mixed‐effects model with ΔQTcF as the dependent variable and time (i.e., post‐baseline timepoint, including the single predose timepoint and all postdose timepoints on days 1, 7, and 14: categorical), treatment (therapeutic and supratherapeutic doses of avacopan, and corresponding placebo), and time‐by‐treatment interaction as fixed effects. The analysis for QTcF was also performed for moxifloxacin versus placebo at predose and postdose timepoints on day 1 and day 15 using similar linear mixed‐effects model. For HR, PR, and QRS intervals, the analysis was based on the change from baseline postdose values (ΔHR, ΔPR, and ΔQRS). In addition, an analysis of categorical outliers was performed for changes in HR, PR, QRS, QTcF, T‐wave morphology, and U‐wave presence.

Statistical methods

All statistical analyses were performed using the statistical software SAS for Windows Version 9.4 (SAS Institute, Cary, NC, USA).

Data sharing statement

Qualified researchers may request data from Amgen clinical studies; complete details are available at http://www.amgen.com/datasharing.

RESULTS

Baseline characteristics

The baseline characteristics and demographics of the study participants are provided in Table 1. A total of 58 participants, including 27 females and 31 males, were enrolled in the study and were randomized to study treatment: 29 participants in Cohort 1, 14 participants in Cohort 2A, and 15 participants in Cohort 2B. All 58 participants completed the study. The mean age of all participants in this study was 40 years and the mean BMI was 26.4 kg/m².

TABLE 1.

Baseline demographics.

	Cohort
	1 (n = 29)	2A (n = 14)	2B (n = 15)	Overall (n = 58)
Gender
Female	13 (45)	6 (43)	8 (53)	27 (47)
Male	16 (55)	8 (57)	7 (47)	31 (53)
Combined	29 (100)	14 (100)	15 (100)	58 (100)
Race
Black or African American	7 (24)	2 (14)	3 (20)	12 (21)
White	22 (76)	9 (64)	12 (80)	43 (74)
White, American Indian/Alaska Native	0	2 (14)	0	2 (3)
White, Black or African American	0	1 (7)	0	1 (2)
Ethnicity
Hispanic or Latino	18 (62)	7 (50)	11 (73)	36 (62)
Not Hispanic or Latino	11 (38)	7 (50)	4 (27)	22 (38)
Age, years a
Mean ± SD	41.7 ± 10.8	37.9 ± 9.4	40.4 ± 11.9	40.4 ± 10.7
Median (range)	40.0 (23–59)	37.0 (22–52)	38.0 (19–60)	38.5 (19–60)
Weight, kg
Mean ± SD	74.2 ± 9.6	74.6 ± 11.9	74.5 ± 12.8	74.4 ± 10.9
Median (range)	76.4 (53.7–92.2)	76.5 (53.3–99.6)	73.8 (54.2–95.4)	76.3 (53.3–99.6)
Height, cm
Mean ± SD	168.6 ± 9.1	168.4 ± 8.8	164.8 ± 9.6	167.6 ± 9.1
Median (range)	170.0 (153–185)	168.0 (151–184)	163.0 (150–180)	168.5 (150–185)
BMI, kg/m2
Mean ± SD	26.1 ± 2.8	26.2 ± 3.3	27.2 ± 2.1	26.4 ± 2.8
Median (range)	26.2 (19.1–29.7)	27.2 (19.5–29.7)	27.2 (24.1–29.7)	26.6 (19.1–29.7)

Note: Data presented as number (%) of participants unless indicated otherwise. Data for age, weight, height, and BMI are rounded off to the first decimal.

Cohort 1: Avacopan 30 mg b.i.d. on days 1 through 7, followed by avacopan 100 mg b.i.d. on days 8 through 14. Moxifloxacin placebo was co‐administered with the morning dose on day 1, and administered alone in the morning on day 15.

Cohort 2A: Avacopan placebo b.i.d. (days 1–14) + moxifloxacin 400 mg was co‐administered with the morning dose on day 1, and moxifloxacin placebo was administered alone in the morning on day 15.

Cohort 2B: Avacopan placebo b.i.d. (days 1–14) + moxifloxacin placebo co‐administered with the morning dose on day 1, and moxifloxacin 400 mg administered alone in the morning on day 15.

Abbreviations: b.i.d., twice daily; BMI, body mass index; SD, standard deviation.

^a Age is derived from birth date to date of first dose.

Effect on heart rate

In the avacopan‐treated participants, the mean change from baseline HR (∆HR) was similar to the ΔHR in the placebo group on days 1, 7, and 14, ranging across all 3 days from −3.0 to 4.2 ms for avacopan and from −2.2 to 4.0 ms for placebo (Figure 2). The mean ∆∆HR across postdose timepoints varied between −3.3 beats per minute (bpm; at 9 h postdose on day 1, 30 mg b.i.d.) and 2.5 bpm (at 2 h postdose on day 14, 100 mg b.i.d.). There were no outliers in terms of HR changes on active treatment except one placebo participant who had an HR above 100 bpm with an increase in ΔHR >25% at one timepoint.

FIGURE 2.

Change from baseline HR (ΔHR) across timepoints for avacopan and placebo. LS mean and 90% CI based on a linear mixed‐effects model: ΔHR = Time + Treatment + Time*Treatment. A compound symmetry covariance structure was used to specify the repeated measures (time within participant). bpm, beats per minute; b.i.d., twice daily; CI, confidence interval; h, hour; HR, heart rate; LS, least square.

Effect on cardiac repolarization: the QT interval

Mean ΔQTcF on avacopan was similar to mean ΔQTcF on placebo on days 1, 7, and 14, ranging across all 3 days from −5.5 to 3.5 ms on avacopan and from −6.9 to 1.4 ms on placebo (Figure 3). The mean ∆∆QTcF across all 3 days ranged from −1.0 ms at 24 h postdose on day 14 (avacopan 100 mg b.i.d.) to 4.9 ms at 5 h postdose on day 14 (avacopan 100 mg b.i.d.; Figure S1). The upper bound of the 90% CI of ΔΔQTcF was below 10 ms at all postdose timepoints on all days. Following dosing with 400 mg moxifloxacin, the mean ΔΔQTcF observed 3 h postdose was 15.8 ms (90% CI: 10.84 to 20.77; Figure S1). There were two participants with QTcF values between 450 and 480 ms (both on avacopan 30 mg b.i.d.) and no participants with QTcF >480 ms. There was one participant with ΔQTcF between 30 and 60 ms on 100 mg avacopan b.i.d. and no participants with ΔQTcF >60 ms. No treatment‐emergent T‐wave morphology changes or U waves were observed.

FIGURE 3.

Change from baseline QTcF (ΔQTcF) across timepoints for avacopan and placebo. LS mean and 90% CI based on a linear mixed‐effects model: ΔQTcF = Time + Treatment + Time*Treatment. A compound symmetry covariance structure was used to specify the repeated measures (time within participant). b.i.d., twice daily; CI, confidence interval; h, hour; LS, least square; QTcF, QT intervals were corrected for heart rate using Fridericia's (cube root) correction.

Concentration‐QTc results

The primary statistical analysis was based on C‐QTc modeling. Based on the AIC values, the full model with both analytes was chosen as the primary model. The relationships between the avacopan, its metabolite M1, and moxifloxacin plasma concentrations and ΔΔQTcF are shown in Figure 4a–c.

FIGURE 4.

Avacopan QTc analysis. Figure shows a scatter plot of observed avacopan (a), metabolite M1 (b), and moxifloxacin (c) plasma concentrations and mean change from baseline in QTcF interval using Fridericia's formula (ΔΔQTcF). The solid red line with dashed red lines denotes the model‐predicted mean ΔΔQTcF with 90% CI. The plotted points denote the pairs of observed drug plasma concentrations and estimated placebo‐adjusted ΔQTcF (ΔΔQTcF) by participants for each positive dose group and placebo group. The individually estimated placebo‐adjusted ΔQTcF _i,k (ΔΔQTcF _i,k ) equals the individual ΔQTcF _i,k for the participant administered with positive drug or placebo at timepoint k minus the estimation of the time effect at timepoint k. The scatter plots for the following are shown here: (a) avacopan (ΔΔQTcF = 1.88 + 0.0005 × avacopan + −0.006 × 161.87, where 161.87 is the arithmetic mean C _max of M1). (b) Metabolite M1 (ΔΔQTcF = 1.88 + −0.006 × M1 + 0.0005 × 515, where 515 is the arithmetic mean C _max of avacopan). (c) Moxifloxacin (ΔΔQTcF = 2.75 + 0.006 × moxifloxacin). b.i.d., twice daily; CI, confidence interval; QTcF, QT intervals were corrected for heart rate using Fridericia's (cube root) correction.

An assessment of the adequacy of the linear mixed‐effects model was given by the goodness of fit plot shown in Figure 5 and Table S1. The estimated population slope of the C‐QTc relationship was shallow but not significant with an estimate of 0.0005 ms per ng/mL (90% CI: −0.00621 to 0.00717) for avacopan and −0.006 ms per ng/mL (90% CI: −0.0371 to 0.0247) for metabolite M1, with a treatment effect‐specific intercept of 1.9 ms (90% CI: −0.10 to 3.85). An effect on ΔΔQTcF exceeding 10 ms can be excluded within the full observed range of plasma concentrations of avacopan and M1, up to ~1220 and ~335 ng/mL, respectively.

FIGURE 5.

Model‐predicted and observed ΔΔQTcF (mean and 90% CI) across deciles of plasma concentrations for avacopan and metabolite M1. The solid black line with gray shaded area denotes the model‐predicted mean ΔΔQTcF with 90% CI. The red filled circles with vertical bars denote the estimated mean placebo‐adjusted ΔQTcF with 90% CI displayed at the associated median plasma concentration within each decile for (a) avacopan, (b) metabolite M1, among which the individually estimated placebo‐adjusted ΔQTcF _i,k (ΔΔQTcF _i,k ) equals the individual ΔQTcF _i,k for participant is administered with avacopan at timepoint k minus the estimation of time effect at timepoint k. The black circle with vertical bars denotes the mean placebo‐adjusted ΔQTcF with 90% CI for placebo at a concentration of 0. The horizontal red line with notches shows the range of concentrations divided into deciles for avacopan. The area between each decile represents the point at which 10% of the data is present; the first notch to second notch denotes the first 10% of the data, the second notch to third notch denotes the 10%–20% of the data and so on. CI, confidence interval; QTcF, QT intervals were corrected for heart rate using Fredericia's (cube root); ΔQTcF, change from baseline in QTcf interval.

The results from the linear mixed‐effects model for positive control moxifloxacin showed a statistically significant, positive slope of the C‐QTc relationship with an estimate of 0.006 ms per ng/mL (90% CI: 0.0033 to 0.0091) (Table S2; Figure S2). The model‐predicted ΔΔQTcF interval at the geometric mean C _max of moxifloxacin (1951 ng/mL) from the model was well above 5 ms (14.8 ms [90% CI: 9.65 to 19.97]), thereby demonstrating assay sensitivity. ⁸

Pharmacokinetic results

The summary of PK parameters for avacopan, its metabolite M1, and moxifloxacin are shown in Table 2, and the plasma concentration–time profiles are shown in Figure S3. Following avacopan 30 mg b.i.d. dosing (therapeutic dose) and 100 mg b.i.d. dosing (supratherapeutic dose), the maximum plasma concentrations of avacopan were observed at 2 h postdose (T _max) on days 1 and 7, and at 3 h postdose on day 14. The geometric mean (geometric CV%) C _max of avacopan following 30 mg b.i.d. and 100 mg b.i.d. was 203.0 (28.5) ng/mL on day 7 and 779.8 (35.0) ng/mL on day 14, respectively. There was a four‐fold increase in C _max following 7 days of b.i.d. dosing with avacopan 100 mg compared to avacopan 30 mg. The T _max of metabolite M1 following avacopan 30 mg b.i.d. and 100 mg b.i.d. dosing was observed at 3 h postdose on days 1 and 7, and 4 h postdose on day 14. The geometric mean (geometric CV%) C _max of metabolite M1 following avacopan 30 mg b.i.d. and 100 mg b.i.d. doses was 83.12 (25.1) ng/mL at day 7 and 231.4 (24.8) ng/mL at day 14.

TABLE 2.

Plasma pharmacokinetic parameters of avacopan, metabolite M1, and moxifloxacin.

	Treatment A: 30 mg avacopan b.i.d. day 1–day 7		Treatment B: 100 mg avacopan b.i.d. day 8–day 14
PK parameters	day 1 (n = 29)	day 7 (n = 29)	day 14 (n = 29)
(a) Avacopan
AUC0–12h (ng*h/mL)	475.7 (40.3)	1216.0 (35.2)	5595.0 (38.6)
C max (ng/mL)	107.7 (33.6)	203.0 (28.5)	779.8 (35.0)
T max (h)	2.000 (1.00, 4.03)	2.005 (1.00, 3.03)	3.000 (2.00, 5.01)
(b) Metabolite M1
AUC0–12h (ng*h/mL)	255.8 (24.0)	697.7 (27.0)	2164 (24.8)
C max (ng/mL)	42.93 (28.2)	83.12 (25.1)	231.4 (24.8)
T max (h)	2.998 (1.99, 4.16)	3.004 (2.00, 4.05)	3.998 (2.00, 5.01)

	Treatment C: 400 mg moxifloxacin 400 mg + avacopan	Treatment F: moxifloxacin 400 mg
	day 1 (n = 14)	day 15 (n = 15)
(c) Moxifloxacin
AUC0–24h (ng*h/mL)	21,090 (20.6)	22,870 (22.9)
C max (ng/mL)	1838 (20.9)	2062 (26.1)
T max (h	2.001 (1.00, 4.04)	2.007 (0.52, 3.06)

Note: Pharmacokinetic parameters of (a) avacopan (b) avacopan metabolite M1 from Cohort 1, Treatment A: Multiple doses of 30 mg avacopan b.i.d. on days 1 through 7 with moxifloxacin placebo co‐administered with the morning dose on day 1; and Cohort 1, Treatment B: Multiple doses of 100 mg avacopan b.i.d. on days 8 through 14 with moxifloxacin placebo administered alone in the morning on day 15.

PK parameters from (C) moxifloxacin from Cohort 2A, Treatment C: Avacopan placebo b.i.d. days 1 through 14 with 400 mg moxifloxacin co‐administered with the morning dose on day 1, and Cohort 2B, Treatment F: Single dose of 400 mg moxifloxacin administered alone in the morning on day 15.

AUCs and C _max are presented as geometric mean and geometric CV%.

T _max is presented as median (minimum, maximum).

Abbreviations: AUC, area under the curve; b.i.d., twice daily; C _max, maximum concentration; PK, pharmacokinetic; T _max, time to reach maximum concentration.

Dosing avacopan 30 mg b.i.d. from day 1 to day 7, the geometric mean AUC values increased by 2.2‐ to 2.6‐times for avacopan and by 2.4‐ to 2.7‐times for its metabolite M1, with the geometric mean of C _max increasing 1.9‐times for both analytes.

After a single dose of 400 mg moxifloxacin, the geometric mean C _max was ~1838 ng/mL and the AUC_0‐24h was ~21,000 ng*h/mL (Figure S3). These PK parameters are similar to those reported in the literature. ⁸

Effect on cardiac conduction

At the studied doses of avacopan, a clinically meaningful effect on cardiac conduction, that is, the PR and QRS intervals was not observed. Mean ∆PR on active treatment generally followed the pattern observed in placebo participants. Mean placebo‐corrected ΔPR (ΔΔPR) varied between −5.7 ms (at 24 h postdose on day 1, 30 mg b.i.d.) and 7.8 ms (at 1 h postdose on day 1, 30 mg b.i.d.). Mean ∆QRS was small and mean ΔΔQRS varied between −1.1 and 1.4 ms across all post‐baseline timepoints. There was one participant on 30 mg b.i.d. with PR above 200 ms with an increase in ΔPR > 25% PR and no QRS outliers.

Safety evaluation

The percentage of participants reporting AEs was 38% following multiple supratherapeutic doses of avacopan (100 mg b.i.d.) and 21% following multiple therapeutic doses of avacopan (30 mg b.i.d.). The most commonly reported AE following avacopan administration was headache (21% of participants). The majority of AEs following avacopan administration were mild in severity and possibly related to avacopan (Table S3A–C). No treatment‐ or dose‐related trends were observed with respect to clinical laboratory, vital sign, ECG, or physical examination safety assessments. There were no deaths or participant discontinuations due to AEs reported in the study. One participant in Cohort 2A (receiving moxifloxacin and avacopan placebo) experienced a serious AE (SAE) of transverse myelitis 31 days following discharge from the study that required hospitalization. The principal investigator considered this SAE to be unlikely related to study drug.

DISCUSSION

The objective of this clinical pharmacology study was to evaluate the effect of avacopan, a C5aR antagonist, following therapeutic (30 mg b.i.d.) and supratherapeutic (100 mg b.i.d.) doses on cardiac repolarization in healthy participants. The results of this study demonstrated that orally administered avacopan at therapeutic dose of 30 mg b.i.d. and supratherapeutic dose of 100 mg b.i.d. did not have a clinically meaningful effect on cardiac repolarization (the QTc interval) or cardiac conduction (the PR and QRS intervals).

This study utilized C‐QTc analysis as the primary statistical method with at least 24 evaluable participants from each of two groups. The selection of this sample size was based on the experience from the International Consortium for Innovation and Quality in Pharmaceutical Development‐Cardiac Safety Research Consortium (IQ‐CSRC) study ¹¹ and observations from 25 QTc studies using C‐QTc analysis. ¹² The simulations conducted showed when using C‐QTc analysis ⁹ as an alternative to more traditional “by‐time point analysis” or Intersection Union Test ¹³ for other drugs, ¹⁴ a sample size of 24 evaluable participants from each of the two groups would provide more than 90% power to exclude that avacopan (and M1) caused more than a 10 ms QTc effect at clinically relevant plasma levels, as shown by the upper bound of the 2‐sided 90% CI of the model‐predicted QTc effect (ΔΔQTcF) at the observed geometric mean C _max of avacopan (and M1) in the study. For avacopan, the calculation assumed a one‐sided 5% significance level with a small underlying effect of 3 ms with avacopan and a standard deviation (SD) of the ΔQTcF of 8 ms.

Based on the phase I studies, co‐administration with itraconazole, a strong CYP3A4 inhibitor, increased avacopan C _max by 87% ³ (1.87‐fold), therefore, maximum exposure expected in patients with GPA and MPA would be ~650 ng/mL. This would represent the worst‐case scenario for the C _max of avacopan since other factors such as food, hepatic or renal impairment had no to minimal effects on avacopan C _max. ³ Therefore, the supratherapeutic dosing regimen of 100 mg b.i.d. administered for 7 days was selected such that it would provide at least a two‐fold increase (approximately C _max = 780 ng/mL on day 7 following 100 mg b.i.d.) in therapeutic C _max concentrations of avacopan following 30 mg b.i.d. dosing in patients with GPA and MPA (~349 ng/mL) ³ and ~1.2‐fold over the maximum exposure expected in the event of CYP3A4 inhibition. Due to formulation limitations, evaluation of doses greater than 100 mg b.i.d. in healthy participants was not possible. The dose of 100 mg b.i.d. has previously been tested and was found to be well tolerated in the phase I clinical study. ² Moxifloxacin 400 mg was selected as the positive control, consistent with the ICH E14 guidance recommendation that the positive control should have an effect on the mean QT/QTc interval of ~5 ms.

The results from the C‐QTc analysis for avacopan demonstrated that there was no clinically relevant impact on QTc prolongation since an effect on ΔΔQTcF exceeding 10 ms was excluded within the full observed range of plasma concentrations of avacopan and M1, up to ~1220 and ~335 ng/mL, respectively, which would provide ~ twofold coverage for exposures expected when co‐administered with a strong CYP3A4 inhibitor. On the other hand, the C‐QTc analysis for moxifloxacin showed a positive and statistically significant slope of 0.006 ms per ng/mL (90% CI: 0.0033–0.0091), with the lower bound of the two‐sided CI of the predicted QT effect (14.8 ms [90% CI: 9.65–19.97]) at the geometric mean peak moxifloxacin concentration (1951 ng/mL) being above 5 ms, thereby demonstrating assay sensitivity, and consistent with prior data. ¹⁴ , ¹⁵ , ¹⁶

In the by‐timepoint analysis, following avacopan doses of 30 mg b.i.d. and 100 mg b.i.d., the mean ∆∆HR was less than 10 bpm across all postdose timepoints, thereby demonstrating that avacopan had no relevant effect on HR.

The mean ΔQTcF with avacopan was similar to ΔQTcF observed following placebo on days 1, 7, and 14. The mean ∆∆QTcF following supratherapeutic dose of 100 mg b.i.d. was 4.9 ms at 5 h on day 14 with the upper bound of the 90% CI below 10 ms at all postdose timepoints on all days, thus demonstrating that avacopan does not cause QTc prolongation following therapeutic or supratherapeutic doses. Following dosing with 400 mg moxifloxacin, a clear increase in mean ΔΔQTcF value was observed with a peak value of 15.8 ms (90% CI: 10.84 to 20.77) at 3 h postdose.

In conclusion, as demonstrated in this positive comparator and controlled QT study, avacopan at the studied doses had no clinically meaningful effects on studied cardiac repolarization, that is, QT/QTc intervals or cardiac conduction, that is, the PR and QRS intervals in healthy participants.

AUTHOR CONTRIBUTIONS

S.M., B.D., H.X., and R.K.O. wrote the manuscript. S.M., B.D., E.T., and P.S. designed the research. D.A. performed the research; K.W. and H.X. analyzed the data. B.D. contributed new analytical tools.

FUNDING INFORMATION

This study was funded by ChemoCentryx, Inc., a wholly owned subsidiary of Amgen, Inc.

CONFLICT OF INTEREST STATEMENT

Funding for this study was provided by ChemoCentryx, Inc., a wholly owned subsidiary of Amgen, Inc. R.K.O. is an employee and shareholder of Amgen Inc. S.M. is a shareholder of Amgen Inc. B.D. is eligible for stock options and owns shares in Clario. All other authors have no relevant funding or compensation to disclose.

Supporting information

Data S1

Miao S, Staehr P, Tai E, et al. A phase I thorough QT/QTc study evaluating therapeutic and supratherapeutic doses of avacopan in healthy participants. Clin Transl Sci. 2024;17:e13878. doi: 10.1111/cts.13878